In October 2021, a pitch went around crypto Twitter that sounded almost too clean: take the voluntary carbon market — an opaque, fragmented system where a "verified" credit could sell for anywhere from $3 to $30 depending on who you asked — and put it on a public blockchain. Every credit traceable, every retirement auditable, every tonne of CO₂ liquid enough to trade like an asset. Within three months, the project backing that pitch had a treasury holding more carbon credits than most mid-size corporations offset in a year.
That project was KlimaDAO, and the infrastructure underneath it was Toucan Protocol. Four years later, both still exist. Neither works the way the pitch said they would. Here's what the on-chain data and the registry that governs the whole system actually show.
How the bridge was supposed to work
Toucan Protocol built what it called a Carbon Bridge. The mechanic was straightforward: someone holding a verified carbon credit on the Verra registry — the largest certifier of voluntary carbon offsets globally — would retire that credit on Verra's books, and Toucan would mint an equivalent token on Polygon. One retired credit, one token, cryptographically linked back to the registry entry so anyone could verify it wasn't double-counted.
Individual tokenized credits (called TCO2) are unique to their specific project and vintage, which makes them clunky to trade. So Toucan pooled them into a fungible token: BCT, or Base Carbon Tonne, which accepts any Verra-certified credit with a 2008-or-later vintage and treats them as interchangeable. Drop a TCO2 into the pool, get one BCT out. That fungibility is what made BCT tradeable on decentralized exchanges in the first place — and, as it turned out, exactly what got the whole model into trouble.
KlimaDAO used BCT as its treasury reserve. The KLIMA token was designed as a carbon-backed currency: every KLIMA minted was matched by roughly one tonne of BCT sitting in the DAO's treasury, following an approach borrowed from OlympusDAO's "protocol-owned liquidity" playbook. Deposit BCT, receive discounted KLIMA, and in theory you're not just buying a token — you're buying a claim on real, retired carbon.
The numbers that mattered were never "grew" — they were "collapsed"
Here's where the marketing and the spreadsheet stop agreeing.
KlimaDAO's treasury did grow fast — the DAO accumulated more than 17 million tokenized carbon credits within its first three months, more than three times Microsoft's entire 2020 annual emissions footprint. That part of the pitch was real.
What happened to the token backed by that treasury was not part of the pitch. KLIMA's price fell from roughly $1 billion in implied treasury value in the spring of 2022 to a token that had lost two-thirds of its value by that same summer. It never recovered the trajectory. By March 2026, KLIMA was trading at $0.041.
BCT tells a similarly blunt story. Across 2025 and 2026, BCT has traded in the 0.08 range with near-zero volume — a token meant to represent a tonne of verified, retired carbon, priced at a level that wouldn't cover a decent cup of coffee. For comparison, the underlying Verra credits it draws from have historically traded anywhere from a few dollars to well over $10 per tonne on the voluntary market. That gap is the whole story in one number: the market stopped believing BCT represented what it claimed to.
The event that actually broke the model
The turning point has a date: May 2023, when Verra announced it would prohibit tokenizing retired credits without the original project developer's explicit consent. KLIMA's price dropped from about $5.68 to a low of $4.74 within hours of the announcement — and that was before the slower, structural damage set in.
The reasoning behind Verra's move is worth sitting with, because it's not a technicality. Retiring a credit is supposed to mean it's used — claimed against someone's emissions, permanently taken out of circulation. Toucan's bridge retired credits on Verra's books specifically so it could re-animate them as a tradeable token on Polygon. Researchers at CarbonPlan, in an analysis pointedly titled "Zombies on the blockchain," documented exactly which credits flowed into BCT — and found the pool had become a dumping ground for the cheapest, oldest, lowest-quality credits available, since anyone holding a weak 2008-vintage credit nobody wanted at full price had every incentive to bridge it into a pool where it traded at par with much stronger projects. Toucan's own bridged tokens represented around 2.4% of Verra's total issued credits by the time the freeze hit — small relative to the whole registry, but large enough to distort pricing in the specific corner of the market it touched.
Verra's decision effectively froze new supply into the BCT pool. Toucan responded with public support for the registry finding "credible pathways" forward; as of March 2026, Verra still had not published a finalized tokenization framework, though the agency has continued engaging with protocol teams including Toucan.
Flowcarbon: the same bet, a bigger name, a token that never launched
Toucan and KlimaDAO weren't alone in this thesis. Flowcarbon — co-founded by former WeWork CEO Adam Neumann — raised $70 million in 2022, including $38 million from a planned token sale, with backing from Andreessen Horowitz's crypto arm. Its token, GNT (Goddess Nature Token), was built on Celo and designed to bundle nature-based credits the same way BCT bundled Verra credits generally.
GNT never launched. Flowcarbon cited "market conditions and resistance from carbon registries" — Verra's tokenization freeze landed on Flowcarbon's plans at almost the same moment it landed on Toucan's — and the company spent the following two years quietly refunding token-sale investors rather than shipping the product it raised money to build. It's a useful data point precisely because Flowcarbon wasn't an amateur operation: serious capital, a serious registry-relationship strategy, and it still couldn't get past the same structural objection that hit Toucan.
What's worth taking seriously here, and what isn't
It would be easy to read all this as "crypto carbon credits were a scam," and that's not quite accurate either. As Wes Ibarra, who covers where blockchain incentives do and don't move real climate outcomes, puts it: "Most 'green crypto' claims fall apart the moment you ask for the methodology. My job is to ask anyway — and to give credit on the rare occasions the math actually holds up." The traceability piece of Toucan's pitch genuinely worked — because Toucan-based retirements are openly recorded in Verra's transaction database, researchers could see exactly which credits became BCTs and evaluate their quality after the fact. That transparency is precisely how the "zombie credit" problem got documented in the first place. A purely off-chain market wouldn't have left that kind of audit trail.
What failed wasn't the traceability — it was the fungibility-over-quality design that let weak credits hide inside a pool priced like strong ones, and a registry relationship that was adversarial rather than built-in from day one. Newer entrants in this space have taken that lesson directly: instead of bridging existing retired credits without the issuer's involvement, some projects are now building tokenization agreements directly with registries like Puro.earth from the outset, so a repeat of Verra's 2023 freeze wouldn't collapse their pipeline the way it collapsed Toucan's.
Wes has covered that shift — from "bridge whatever credits you can find" to "get the registry's buy-in first" — as part of his ongoing reporting on regenerative finance projects, on his author page.
The lesson buried in a $0.04 token
The Toucan/KlimaDAO story isn't really about carbon, and it isn't really about crypto either — it's about what happens when a tokenomics model optimizes for liquidity before it secures quality. A pool that treats every input as interchangeable will always attract the worst inputs first, whether the underlying asset is a carbon credit, a mortgage, or anything else someone wants to offload at par. BCT and KLIMA's current prices aren't a verdict on whether carbon markets belong on a blockchain. They're a verdict on one specific way of building that bridge — and a fairly expensive lesson, at $70 million-plus across just these few projects, on why registries tend to insist on knowing who's tokenizing their credits before they let it happen.